Common Voice

Live Deployment: Commvoice

Data

The data for this project is sourced from Common Voice, which is a crowdsourcing project started by Mozilla to create a free database for speech recognition software. The project is supported by volunteers who record sample sentences with a microphone and review recordings of other users. The transcribed sentences will be collected in a voice database available under the public domain license CC0. This license ensures that developers can use the database for voice-to-text applications without restrictions or costs. Common Voice appeared as a response to the language assistants of large companies such as Amazon Echo, Siri or Google Assistant

Overview

The goal for this project is to create an end to end machine learning appliacation that records and processes audio in real time and stream prediction via a socket API. There's a 1 second delay delay between the audio recording and the output prediction.

The application generates prediction in 3 categories: Gender, Age and Country of Origin.

Todo:

Traing and implement models for Country and Age

Getting Started:

Train Model

Modify the Data Directory to your own direcory

class DataDirectory:
    DATA_DIR = r"C:\Users\ander\Documents\common-voice-data"
    DEV_DIR = r"C:\Users\ander\Documents\common-voice-dev"
    CLIPS_DIR = r"C:\Users\ander\Documents\common-voice-data\clips"

Run machine learning pipeline

python run_pipeline.py

Deploy Model

Windows Machine

pip3 install -r requirements.txt
python3 run_app.py

Linux Install the below on the server prior to running the docker images

Step 1: Install Docker and Docker Compose

# Install Docker compose 
sudo curl -L "https://github.com/docker/compose/releases/download/1.28.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose
sudo chmod +x /usr/local/bin/docker-compose

sudo ln -s /usr/local/bin/docker-compose /usr/bin/docker-compose

# Install Docker 
sudo apt-get update

sudo apt-get install \
    apt-transport-https \
    ca-certificates \
    curl \
    gnupg-agent \
    software-properties-common

curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -

sudo apt-key fingerprint 0EBFCD88

sudo add-apt-repository \
   "deb [arch=amd64] https://download.docker.com/linux/ubuntu \
   $(lsb_release -cs) \
   stable"

sudo apt-get update
sudo apt-get install docker-ce docker-ce-cli containerd.io

Step 2: Install Nginx

# Install NGINX 
sudo apt install certbot python3-certbot-nginx
sudo nano /etc/nginx/sites-available/commvoice.me
...
server_name WEBSITE_NAME WEBSITE_NAME;
...

Step 3: Install Certbot

# Install Certbot 
sudo certbot --nginx -d commvoice.me -d www.commvoice.me

# Install Dhparam 
openssl dhparam -out /etc/nginx/dhparam.pem 2048

# Install Certbot Auto Renew 
systemctl status certbot.timer

Step 4: Install and audio Drive Enabled a snd-aloop modules

modprobe snd-aloop

The below devices should have been added to you dev/snd directory.

ls /dev/snd/

    -  pcmC0D0c
    -  pcmC0D0p
    -  pcmC0D1c

Step 5: Run Docker Compose

docker-commpose up --build